CIPD - Basic pay settlements: data sources

Pay and reward

Revised June 2009

This factsheet gives introductory guidance. It:

reviews the main publicly available sources of pay settlement data in the UK.

Overview of the main sources

The main providers of publicly available pay settlement data are the EEF The Manufacturers’ Organisation (EEF), Incomes Data Services (IDS), Industrial Relations Services (IRS), and the Labour Research Department (LRD). The Office of Manpower Economics (OME) ceased its data collection exercise in June 2005. The Confederation of British Industry (CBI) suspended its pay databank in December 2003 and is unlikely to recommence collecting and issuing settlement data in the immediate future.

Basic data derived from their surveys are often available on the websites of these organisations, but more detailed analysis can generally only be obtained by subscribing to the relevant publications or via membership of the organisation concerned.

The main features of the various databases are set out in the table below.

Comparisons of main settlement data sources

	EEF	IDS	IRS	LRD
Structured sample?	no	no	no	no
Approx no. of awards	1,500	1,000	1,200	1,000
Settlement source	employers	employers and unions	employers and unions	unions
Basic/Total pay	basic	basic	basic	basic
Public sector included?	no	yes	yes	yes
Manual/ non-manual identified separately?	no	no	yes	no
Median/ Average/ Range	all	median/ range	all	median

There are other sources of data on pay movements supplied by:

Management consultancies and organisations such as Croner Reward. Some of these can be expensive to access, or are only available to participants, and the results are largely based on changes between base or earnings levels year-on-year, as distinct from assessments of pay awards per se.
The Annual Survey of Hours and Earnings (which has replaced the New Earnings Survey) produced by Office for National Statistics and available on their website gives changes to basic pay and earnings levels.
The quarterly Inflation Report of the Bank of England also contains some analysis of average settlement levels based on the Bank’s own assessment of data drawn from its regional agents and the main commentators mentioned above.
CIPD collects data on expected pay increases in our quarterly Labour market outlook. We ask respondents when their next pay review is due and what they predict the size of the pay rise will be. The sample size is around 1,000 employers and pay data is analysed pay data by economic sector, the median and the interquartile range.
- Go to the latest Labour market outlook
The Society for Human Resource Management (SHRM) run a similar survey which gives pay data for the USA.

Explanation of terms

The main terms used in this factsheet are:

Basic pay: all the main commentators ‘price’ settlements in respect of their effect on basic pay rates, as defined by them. That is, excluding the effect of any additional payments over and above the basic increase, such as consolidation, incremental rises, merit pay, bonuses, and overtime and shift allowances.
Mean: (or average) is the sum of all the relevant percentage basic awards divided by the total of all the observations. A few, unrepresentative, unusually high or low awards can therefore influence the mean.
Median: the mid-point of the range of observations, that is the point where fifty percent of settlements are either at or below, or at and above, the figure. It is less susceptible than the mean to influence by outlier awards.
Mode: is the most frequently occurring value in a set of observations. No commentator publishes a modal figure, but it is often similar to the median settlement level.
Quartiles: are obtained by ranking all the relevant observations. The point where 25% of observations are at or below the figure is the lower quartile and the point where 25% are at or above the figure is the upper quartile. The spread between these two points – the inter-quartile range – gives an indication of the extent to which settlements ‘bunch’ around the median.

Information on how such data is used in the annual pay review process can be viewed at our factsheet on this topic.

Go to our factsheet on the Salary review process

Pay and reward concepts generally are covered in our overview factsheet on pay and reward.

Go to our factsheet on Pay and reward: an overview

EEF The Manufacturers’ Organisation

The EEF has been collecting and publishing pay settlement data since 1980. The individual Engineering Employers’ Associations collect the data regionally. They identify the settlement date for each member and contact the employer a month before the settlement is due following this up, if necessary, a month after the due date. Each Association except Northern Ireland produces its own (unpublished) regional figures, but each also sends the individual settlement data to the EEF centrally for analysis to produce a national figure. EEF staff will also ‘sore thumb’ the data to ensure there are no glaring inconsistencies. The data are published monthly in arrears in the EEF pay bulletin, usually following release by the Office for National Statistics of that month’s retail prices data. The analysis is available on the EEF website to members only, but the results are often reviewed in publications such as IDS Pay Report¹ and the IRS Employment Review¹.

The data collected cover:

the effective date of the review
percentage increase in basic pay (excluding bonuses)
number of employees covered by the settlement
the period of time the settlement covers
the latest award if known (or forecast if not yet agreed) and equivalent award a year previously.

Data are collected for ‘manual’, ‘staff’ or ‘both’, but no separate figures are published by these broad categories. Where an award contains more than one settlement figure, that applying to the largest group of employees is included in the analysis. Settlements for less than 12 months are not included in the data. Performance payments are excluded.

Some 1500 settlements are reviewed covering around 240,000 employees. About a fifth of the EEF's membership takes part in the survey, mostly drawn from engineering but also from manufacturing more widely. The sample is not structured in any way to delineate between different sectors of the industry.

The main figures produced are the percentage average increase for the latest month and the latest three months, along with the percentage of settlements that were either deferrals or pay freezes. Deferred settlements are not included in the aggregate figure. The monthly and three-monthly averages are also represented graphically, as is the proportion of pay freezes. Medians and quartiles are provided in histogram form.

Incomes Data Services

Incomes Data Services (IDS) began collecting pay data in 1966. Until quite recently IDS restricted its analysis of aggregate pay data to its Pay Chart which plots settlements against the Retail Prices Index, and to histograms showing the percentage of pay settlements in percentage ranges. This reflected IDS’s view that its role was to report pay changes in context, leaving subscribers to draw their own conclusion as to trends. Introducing the Pay Chart in 1980, IDS said that the variety of settlements made ‘generalisations almost meaningless’ and that ‘true comparisons are only valid between specific settlements’.

These are views that IDS continues to hold. However, subscriber pressure led to:

the introduction of statistical analysis giving quarterly data on pay settlement medians and the inter-quartile range
more recently, to the provision of data on a three-month rolling basis.

Textual analysis also gives a breakdown of any ‘lumpiness’ in the distribution, previously captured in the histograms, and of the respective medians in the public and private sectors. The Pay Chart is retained. Data are published once a month in IDS Pay Report¹.

IDS analyses just over 1,000 settlements covering some 9 million employees. Seventy-five percent of these awards are company or industry-level settlements, 13% are in the public sector and the remainder in the not-for-profit sector. IDS claims a fairly comprehensive coverage of large organisations in all sectors, and a good number of small and medium size organisations, with particularly good coverage in industries such as retail, pubs and restaurants, engineering, electronics, aerospace, chemicals, and transport.

The percentage figure recorded for each settlement is the increase to basic pay. Bonuses or lump sum payments are not included in the aggregate data, although information on these payments is shown in the table of latest settlements that accompanies the analysis. For settlements where the percentage increase varies for different employees (for example, to reflect individual performance), the figure recorded is the average increase where known, or alternatively the increase received by most employees, or the paybill rise. In the case of staged awards only the first stage is reported and there is no attempt to annualise pay deals that are for other than 12 months.

IDS maintains a database of pay settlements that indicates when settlements are due. Researchers contact HR and reward managers to talk through awards and/or any continuing negotiations. Employers’ organisations and trade unions are also contacted for discussions about general pay trends as well as industry-specific negotiations. The pay database is augmented by regular surveys of pay and conditions in specific sectors and industries, and for specific occupations.

Industrial Relations Services

Industrial Relations Services (IRS) has been collecting pay data since 1971. It established its pay databank in 1984. This was updated in 1993 to allow for more complex analysis, and again in 2002 to include more comprehensive details on terms and conditions. It now contains details on around 4000 organisations, and 23,000 separate pay awards.

Data are collected directly from organisations by IRS’s research staff using a ‘settlement calendar’ to identify customary settlement dates. Generally, details are collected from management, supplemented by additional information from trade unions. Where possible hard copies of pay agreements are sought.

In terms of the pay databank the key details collected are:

the percentage increase for the lowest adult grade
the effective award date
the number of employees covered
the duration of the award.

Information is also collected on:

terms and conditions such as basic hours and holidays
merit pay and bonuses
premium payments
pensions and all other benefits.

This information forms the basis of separate subject analysis in IRS publications. Similarly, IRS produces separate analyses of pay trends in particular industries and sectors.

In terms of the pay databank analysis, aggregated data are published in respect of 1200 or so awards. Around 1000 awards each year are included in the monthly pay chart. It is important to note that the key figure recorded is the percentage increase in the lowest adult basic rate. It therefore excludes other pay elements such as merit and performance pay, bonuses, overtime, shift and other allowances. Some 300 merit-only awards are collected each year and are separately reviewed on a monthly basis to produce a figure based on the median percentage merit ‘pot’.

Data are published monthly in IRS Employment Review ¹ in the form of charts and tables showing the three-monthly and twelve-monthly means, medians, upper and lower quartile figures for the whole economy, private and public sectors by number of pay reviews. (The whole economy median is also calculated by number of employees, but a few large awards, particularly for public sector groups, can heavily influence these figures and the analysis by review is preferable).

The private sector is broken down further into manufacturing and services, and into manual and non-manual. Overall, 91% of settlements are from the private sector of which 44% are in manufacturing and 54% in services, and 47% cover manual workers and 42% cover non-manual workers (the remainder may cover both, but this is not specified)). 9% of settlements are in the public sector.

In terms of employees covered, some 14% are employed in manufacturing and production, 32% in private sector services, and 54% in the public sector. Within the broad sectors there are clear biases as certain industries are over-represented. IRS acknowledges this point, noting that information is provided by employers on a voluntary basis, and achieving a representative sample is a challenge. However, it does seek to redress this to some extent by gathering information across the range of organisational size.

For pay awards that run for more or less than 12 months, the figures are annualised. That is, an award of 3.5% due to run for 18 months is recalculated as being worth 2.3% on annualised basis. Pay freezes for the full twelve months are entered as a ‘zero’. A pay ‘pause’ or ‘deferral’ followed by a percentage uplift is annualised to give one figure for the year, for example, a six-month deferral followed by a 2.5% uplift would be entered as 1.25% for the year.

Labour Research Department

The Labour Research Department (LRD) is a research body funded by several trade unions and from commercial publishing activities. It has no direct connection with the Labour Party.

The LRD pay databank dates from 1983 and now covers 1,000 settlements. Analysis is published monthly in Workplace Report¹, and the latest basic median figures are published monthly on the website. Median figures are presented for the whole economy and public and private sectors separately, by number of agreements and by number of employees covered. The data are prepared on a three-monthly and a twelve-monthly basis, and in respect of the latest pay round to date, that is, for the period since 1 August in any twelve month period.

Trade union negotiators at national or local level are the main sources of data, and the sample is therefore heavily biased towards collectively bargained awards in unionised sectors of the economy.

LRD adopts three standard ‘yardsticks’ for valuing awards:

Like IRS, LRD costs an award as the percentage uplift in the lowest basic rate or scale minimum (avoiding basing this on age-related, probationary or similar rates). This often equates to the general award applicable to all employees, but it can be affected by increases geared towards the lowest paid, or consolidations of bonuses etc into the basic rate.
Where there is some unrepresentative effect on the award, for example, higher awards in the lower grades, LRD also tries to recode the ‘average’ pay rise across all the grades covered by the agreement.
Where the overall cost of the settlement is expressed as a percentage rise in the paybill, LRD also seeks to capture that. This would include elements of total pay and would probably be a higher figure than the basic award.

Where agreements are staged, the reported figure refers to the overall increase in the lowest basic rate during the pay year in question. Individual performance-related increases, or rises based on progression, are not included.

Annual distribution of pay settlements

Pay settlements are not evenly distributed through the year and they tend to bunch in January and April, which between them account for nearly half of all settlements. This possibly reflects traditional financial reporting years in the public and private sectors, (there is also a smaller ‘blip’ in July). It does mean, however, that the period from August to December is ‘fallow’, with only around a fifth of settlements in total being concluded during those five months. This should be borne in mind when interpreting three-monthly pay settlement data, particularly towards the end of the year.

Long-term pay deals

The different commentators have different approaches to valuing pay awards that last for more (or less) than a year. To an extent this has been complicated by the increasing use of longer-term pay deals in both the public and private sectors. This also means that later stages of awards can give a slightly false impression of the generality of awards that are being reached at the time, and this needs to be considered when reviewing the pay scene.

In May 2004, IDS noted a large increase in the use of long-term deals². It says that around a fifth of pay increases were paid under long term deals, over double the incidence typical in the 1990s, and the highest level they have recorded since the 1960s. IDS argued that the increase reflects both a large growth in such deals in the public sector, and an upturn in interest in the private sector. Overall, of the long-term deals IDS analyses, two-thirds are in the private sector and a third in the public sector. It is ambivalent as to whether such deals will increase, given the lack of ‘sweeteners’ in the public sector which had previously made them quite attractive to staff, and their concentration in a few industries in the private sector. In October 2006, IDS suggested that the popularity of such deals could decline with inflation forecast to rise.

Conclusion

All the main commentators publish data only on changes to base rates, as defined by them. IRS publish data specific to non-manual employees in the private sector. The other providers do not undertake this breakdown.

Each of the surveys has its weaknesses and none can be said to be definitive.

Most are not based on statistical samples of the economy, and are dependent on a self-selecting group of respondents.
Individual settlements may be open to different interpretations and their inclusion, or the actual figure used, can sometimes depend on the views of the commentators.
There are also some concerns about whether flexible approaches to pay are adequately reflected in the data.

Overall, however, despite the differences between them, the data do give a reasonably consistent picture of the state of basic pay awards in the UK economy.

Useful contacts

Industrial Relations Services

Labour Research Department

Office for National Statistics

References

CIPD Library has the journals IDS Pay Report, IRS Employment Review, Workplace Report which are available to members in the Library.
Understanding reward: long term pay deals are increasingly important. (2004) IDS Pay Report. No 905, May. pp8-10.

Basic pay settlements: data sources